AMD Radeon RX 9070 vs Intel Arc B570 Comparison

AMD
RADEON

AMD Radeon RX 9070

CORE STATE Navi 48
VRAM 16 GB
CLOCK SPEED 2520 MHz
TDP 220 W
BUS WIDTH 256 bit
ARCHITECTURE RDNA 4.0
nm
PROCESS 4 nm
LAUNCH DATE 2025
VS
Intel
GPU

Arc B570

CORE STATE BMG-G21
VRAM 10 GB
CLOCK SPEED 2500 MHz
TDP 150 W
BUS WIDTH 160 bit
ARCHITECTURE Xe2-HPG
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
6,290
2,649
geekbench_opencl
131,539
83,514
geekbench_vulkan
58,705
96,844
passmark_directx_10
141
65
passmark_directx_11
281
118
passmark_directx_12
74
72
passmark_directx_9
343
164
passmark_g2d
1,280
661
passmark_g3d
25,381
14,195
passmark_gpu_compute
14,737
7,281

Analysis: AMD Radeon RX 9070 vs Intel Arc B570

# AMD Radeon RX 9070 vs Intel Arc B570: Benchmark Database Analysis

The database places the AMD Radeon RX 9070 and the Intel Arc B570 in adjacent performance tiers. Both are modern discrete graphics solutions, but the recorded benchmark data shows a decisive split in their capabilities. The AMD Radeon RX 9070 holds a substantial lead in the majority of tested workloads, while the Intel Arc B570 secures one notable victory in a specific API workload. This analysis walks through the head-to-head results, the use-case split, and the architectural differences that explain the recorded outcomes.

Head-to-Head Benchmarks

The database includes ten head-to-head benchmark comparisons between the AMD Radeon RX 9070 and the Intel Arc B570. The AMD card wins nine of these ten comparisons, with the Intel card taking a single victory. The magnitude of the AMD wins varies considerably by test type, from a narrow margin to a dominant lead.

The largest single victory for the AMD Radeon RX 9070 comes in the 3DMark Steel Nomad DX12 test. Here the AMD card scores 6290 points against 2649 points for the Intel Arc B570. That is a performance advantage of 137.4 percent. This represents a near textural doubling of performance in a modern DirectX 12 workload.

In the Geekbench OpenCL test, the AMD Radeon RX 9070 records a score of 131539, while the Intel Arc B570 scores 83514. The AMD advantage here is 57.5 percent. This is a compute-focused test, and the result indicates that the RX 9070's raw throughput in general purpose compute tasks is markedly higher.

The Passmark DirectX 11 test shows a similarly large gap. The AMD card scores 281 versus 118 for the Intel card, a delta of 138.1 percent. In the PassMark DirectX 10 test, the AMD card again wins by a wide margin: 141 versus 65, a 116.9 percent advantage. The older DirectX 9 workload also favors the AMD card: 343 versus 164, which is 109.1 percent ahead.

In 2D graphics, the gap widens further. The PassMark G2D test shows the AMD Radeon RX 9070 at 1280 points, while the Intel Arc B570 scores 661, a difference of 93.6 percent. The PassMark 3D test (PassMark G3D) sees the AMD card score 25381, the Intel card score 14195, a 78.8 percent lead. Finally, in the PassMark GPU Compute test, the AMD card scores 14737 against the Intel's 7281, a 102.4 percent advantage.

The narrowest AMD win is in the PassMark DirectX 12 test, where the scores are 74 for the RX 9070 and 72 for the Arc B570. The advantage is just 2.8 percent, making this a near tie. The other PassMark DirectX tests follow a similar pattern. The 3DMark Steel Nomad was not the only workload where the AMD card dominates.

The single Intel win is in the Geekbench Vulkan test. The Intel Arc B570 scores 96844, while the AMD Radeon RX 9070 scores 58705. That is a -39.4 percent difference in favor of the Intel card. This is a substantial victory for the Intel architecture in a cross-platform graphics API.

Where Each One Wins

Looking at the ten head-to-head results, the AMD Radeon RX 9070 is the clear winner in compute-heavy and traditional rasterization workloads. It wins in 3DMark Steel Nomad DX12, both Geekbench OpenCL tests, and all four PassMark DirectX tests (9, 10, 11, and 12). It also wins the PassMark 2D test and the PassMark GPU compute test.

The Intel Arc B570 wins in exactly one category: the Geekbench Vulkan test. This suggests the Intel architecture has a particular strength in the Vulkan API, a cross-platform graphics and compute standard. The margin of victory here is substantial, the Intel card outperforms the AMD by 39.4 percent.

For the majority of users, the AMD Radeon RX 9070 will be the stronger pick. It offers a massive lead in DirectX workloads, which are the dominant gaming API in the Windows ecosystem. The data shows consistent wins across every DirectX version tested. The AMD card also holds a significant advantage in OpenCL compute workloads, which points to GPGPU and productivity tasks.

The Intel Arc B570, however, is the better choice for users who rely heavily on Vulkan. The database shows it is 39.4 percent faster than the AMD in the Geekbench Vulkan test. This could be of interest to Linux users, or those who run workloads that are specifically optimized for Vulkan.

FAQ

Q: Which card is faster in the 3DMark Steel Nomad DX12 test?

A: The AMD Radeon RX 9070 is faster. It scores 6290 points against 2649 points for the Intel Arc B570, which is 137.4 percent ahead in this workload.

Q: How does the AMD Radeon RX 9070 compare to the Intel Arc B570 in the PassMark DirectX 11 test?

A: The AMD card records a 138.1 percent advantage. The RX 9070 scores 281, the Arc B570 scores 118.

Q: Which card has the better result in the Geekbench Vulkan benchmark?

A: The Intel Arc B570 wins this test. Its score of 96844 is 39.4 percent higher than the AMD Radeon RX 9070's 58705.

Q: In which benchmarks does the Intel Arc B570 beat the AMD Radeon RX 9070?

A: The Intel card wins exactly one of the ten head-to-head tests: the Geekbench Vulkan result. The AMD card wins in all other nine tests.

Q: What is the difference in the PassMark GPU Compute test results?

A: The AMD Radeon RX 9070 scores 14737, the Intel Arc B570 scores 7281. The AMD is 102.4 percent ahead in this compute-focused workload.

Specification Differences

The two cards differ in several fundamental specifications that explain the benchmark splits. The AMD Radeon RX 9070 is built on the Navi 48 chip, using the RDNA 4.0 architecture, and is manufactured on a 4 nm process at TSMC. The Intel Arc B570 uses the BMG-G21 chip, with the Xe2-HPG architecture, manufactured on a 5 nm process at TSMC.

Memory configurations are notably different. The AMD card has 16 GB of GDDR6 memory on a 256 bit bus, with a memory bandwidth of 644.6 GB/s. The Intel Arc has 10 GB of GDDR6 on a 160 bit bus, with a memory bandwidth of 380.0 GB/s.

The AMD card has 3584 shading units, 224 texture mapping units, 128 ROPs, and 56 ray tracing cores. The Intel Arc has 2304 shading units, 144 TMUs, 80 ROPs, and 18 ray tracing cores. The RX 9070 does not have tensor cores, nor does the Arc B570.

The clock behavior also differs. The AMD has a base clock of 1330 MHz, a boost of 2520 MHz, and a game clock of 2070 MHz. The memory clock is 2518 MHz, which is 20.1 Gbps effective. The Intel Arc has a base clock of 2500 MHz and a boost of 2500 MHz. The memory clock is 2375 MHz, 19 Gbps effective.

In raw throughput, the AMD has a peak FP32 of 36.13 TFLOPS, and an FP16 of 36.13 TFLOPS (1:1). The Intel Arc has an FP32 of 11.52 TFLOPS and FP16 of 23.04 TFLOPS (2:1). The pixel rate for the AMD is 322.6 GPixel/s, texture rate is 564.5 GTexel/s. The Intel rates are 200.0 GPixel/s and 360.0 GTexel/s.

Power and board requirements also differ. The AMD card has a total board power of 220 W, with 2x 8-pin power connectors, and requires a 550 W power supply. The Intel card has a 150 W TDP, uses 1x 8-pin, and requires a 450 W supply.

For API support, the AMD lists DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Intel Arc lists DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Architecture Differences

The architectural split between these two cards is significant. AMD moves to the RDNA 4.0 architecture, which is a major update from the previous generation and is fabricated on a 4 nm node. Intel's Arc B570 uses the Xe2-HPG architecture, which is the second generation of the Intel discrete graphics architecture.

The compute resource allocation differs. The AMD has more shading units (3584 vs 2304), more texture mapping units (224 vs 144), and more ROPs (128 vs 80). It also has more ray tracing cores (56 vs 18). The AMD card therefore has a clear architectural resource advantage.

The memory subsystem also favors the AMD card, with a 256-bit bus versus 160-bit, and more memory (64 GB vs 10 GB). These resources align with the benchmark results, where the AMD card wins in most tests that are limited by shader throughput, memory bandwidth, or compute.

The Intel card's one win in Vulkan may be a sign of a particular driver or architecture advantage for that API. It is not enough to offset the broad lead of the AMD in the recorded data.

The Verdict

From the recorded data, the AMD Radeon RX 9070 is the stronger card for the majority of workloads. It holds a 137.4 percent lead in the 3DMark Steel Nomad DX12 test, a 57.5 percent lead in Geekbench OpenCL, and a 102.4 percent lead in PassMark GPU Compute. The nine test wins cover the primary gaming APIs, the modern general-purpose compute APIs, and legacy D3D workloads. It also has a higher absolute benchmark average: the database records an average benchmark score of 23877 for the AMD, versus 20556 for the Intel, a 16.2 percent difference.

The Intel Arc B570 is the right choice only for the specific Vulkan workload. The Geekbench Vulkan test shows it 39.4 percent faster than the AMD card. For the rest, the AMD leads by a large margin, sometimes more than double the performance. The data is clear: for general gaming and compute tasks, the AMD Radeon RX 9070 is the safer pick. The Intel Arc B570 is a niche pick for Vulkan-focused users or Linux environments where Vulkan is a primary API.

DETAILED SPECIFICATIONS

SPECIFICATION
RX 9070
B570
Core Specs
Shading Units
3,584
2,304 -35.7%
Shaders
3,584
2,304 -35.7%
TMUs
224
144 -35.7%
ROPs
128
80 -37.5%
Compute Units
56
Execution Units
18
Clocks
Base Clock
1330 MHz
2500 MHz
Boost Clock
2520 MHz
2500 MHz
Game Clock
2070 MHz
Memory Clock
2518 MHz 20.1 Gbps effective
2375 MHz 19 Gbps effective
Memory
Memory Size
16 GB
10 GB
VRAM (MB)
16,384
10,240 -37.5%
Memory Type
GDDR6
GDDR6
Memory Bus
256 bit
160 bit
Bandwidth
644.6 GB/s
380.0 GB/s
Cache
L1 Cache
256 KB (per EU)
L2 Cache
8 MB
13.5 MB
L3 Cache
64 MB
L0 Cache
32 KB per WGP
Performance
Pixel Rate
322.6 GPixel/s
200.0 GPixel/s
Texture Rate
564.5 GTexel/s
360.0 GTexel/s
FP32 (TFLOPS)
36.13 TFLOPS
11.52 TFLOPS
FP64 (TFLOPS)
1,129.0 GFLOPS (1:32)
720.0 GFLOPS (1:16)
FP16 (TFLOPS)
36.13 TFLOPS (1:1)
23.04 TFLOPS (2:1)
AI/RT
RT Cores
56
18 -67.9%
XMX Cores
144
Matrix Cores
112
Power
TDP
220 W
150 W
TDP (W)
220
150 -31.8%
Suggested PSU
550 W
450 W
Power Connectors
2x 8-pin
1x 8-pin
Architecture
Architecture
RDNA 4.0
Xe2-HPG
GPU Name
Navi 48
BMG-G21
Generation
Navi IV (RX 9000)
Battlemage (Arc 5)
Process Size
4 nm
5 nm
Transistors
53,900 million
19,600 million
Die Size
357 mm²
272 mm²
Foundry
TSMC
TSMC
Density
151.0M / mm²
72.1M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
2.2
3.0
Shader Model
6.9
6.6
Physical
Slot Width
Dual-slot
Dual-slot
Length
272 mm 10.7 inches
Height
115 mm 4.5 inches
Outputs
1x HDMI 2.1b3x DisplayPort 2.1a
1x HDMI 2.1a3x DisplayPort 2.1
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x8
Other
Launch Price
549 USD
219 USD
Production
Active
Active
Predecessor
Navi III
Alchemist
View Radeon RX 9070 Details View Arc B570 Details